Searcharxiv⌕ Search

arXiv subjects

Gaël Richard

Publications and source records attributed to Gaël Richard.

At least 19 recordsLinked to original sources
arXiv2026

The Internet Archive Music Dataset

We introduce the Internet Archive Music Dataset (IAMD), a large-scale collection of captioned music segments derived from the Internet Archive. To the best of our knowledge, IAMD constitutes the largest publicly available music-caption dataset to date with over 34,000 hours of audio, providing a valuable benchmark for training and evaluating music understanding and generative models. The dataset is built from content declared to be distributed under Creative Commons licenses, and cross-referencing with MusicBrainz is done to improve license information reliability. To annotate IAMD, we present an automatic captioning pipeline that augments base captions produced by an audio-language model (ALM) with textual metadata sourced from the Internet Archive and imputed metadata obtained using audio classification models. Caption quality is assessed objectively and subjectively, and results indicate that the annotation pipeline is reliable and does not degrade caption quality with scaling.

cs.SD↗
arXiv2026

SonicCaps: Large-Scale Diverse and Fine-Grained Captioning for Improved Audio-Retrieval

Recent advances in audio-language modeling have been driven by large-scale audio captioning datasets. However, existing datasets remain limited by low semantic diversity, generic descriptions lacking acoustic details, and one-to-one audio-caption mappings that poorly reflect the inherent ambiguity of auditory perception. We introduce SonicCaps, a large-scale audio captioning dataset comprising ~15M captions paired with ~700k audio clips, generated using a multi-modal large language model (Qwen3-Omni) conditioned on both audio and text. To explicitly promote diversity, we generate around 24 captions per audio via structured prompt engineering and few- shot generation, spanning main descriptions, rephrased variants (verbosity, style) and semantic tags. Human evaluation shows that SonicCaps is rated significantly higher than existing captioning datasets, with fine-grained analyses indicating that our captions are perceived as more descriptive and precise, which strongly correlates with quality judgments. Finally, training CLAP models on SonicCaps with a multi-caption sampling strategy consistently improves audio retrieval and zero-shot classification, with stronger generalization across public and commercial benchmarks. We release both SonicCaps and two specialized CLAP models on hugging face: https://huggingface.co/datasets/Zineb/SonicCaps.

cs.SD↗
arXiv2026

Unlocking Spatial Grounding in Large Audio-Visual Retrieval models

Weak supervision sets a practical regime for audio-visual sound source localization as dense spatial annotations are costly to obtain at scale. The task, however, remains challenging, as models must locate sound sources from temporally aligned audio-visual data without pixel-level supervision. Recent large-scale audio-visual retrieval models, trained at unprecedented scale, encode rich multimodal structure. We show their latent representations, though optimized for global alignment, can nonetheless enable fine-grained spatial grounding. While spatial detail is progressively lost in the upper layers of retrieval backbones due to global pooling, intermediate visual tokens retain highly structured spatial information. To exploit this, we introduce LAIP (\emph{Localization via Audio-Informed Pooling}), a framework that employs a lightweight \emph{Audio-informed Spatial Pooling} (AiSP) to replace the standard global aggregation module. By using frame-aligned audio to query intermediate visual tokens, LAIP recovers localized spatial information that is otherwise discarded by the frozen retrieval pipeline. Our approach achieves state-of-the-art performance on AVSBench and AVATAR, nearly doubling previous results on the latter. These findings prove that accurate localization does not need to be learned from scratch; instead, it can be unlocked from existing retrieval representations, providing a unified path for both retrieval and localization tasks.

cs.IR↗
arXiv2026

Multiple Choice Learning of Low-Rank Adapters for Language Modeling

We propose LoRA-MCL, a training scheme that extends next-token prediction in language models with a method designed to decode diverse, plausible sentence continuations at inference time. Traditional language modeling is an intrinsically ill-posed problem: given a context, multiple futures may be equally plausible. Our approach leverages Multiple Choice Learning (MCL) and the winner-takes-all loss to efficiently handle ambiguity through Low-Rank Adaptation. We provide a theoretical interpretation of applying MCL to language modeling, assuming the data is generated from a mixture of distributions. We illustrate the proposed approach using mixtures of Markov chains. We then demonstrate with experiments on audio and visual captioning, as well as machine translation, that our method achieves high diversity and relevance in generated outputs. We release the code for applying LoRA-MCL to a wide range of language models.

cs.LG↗
arXiv2026

Taylor-SWFT: fast discrete Statistical Wave Field Theory using Taylor expansion for late reverberation Work under review

Dynamic room acoustic simulation aims to render the acoustic effects of an environment in real time while accounting for potentially moving sources and receivers. In this context, the efficient synthesis of the long-term room response, also known as late reverberation, remains challenging because of the intricate relationship between room geometry and acoustic behavior. This paper introduces Taylor-SWFT, an efficient implementation of key results from Statistical Wave Field Theory (SWFT) for the geometry-aware dynamic synthesis of late reverberation. The method is evaluated on the Benchmark for Room Acoustical Simulation (BRAS) and achieves competitive performance compared with classical approaches, while substantially reducing computational cost.

physics.class-ph↗
arXiv2026

U-DREAM: Unsupervised Dereverberation guided by a Reverberation Model

This paper explores the outcome of training state-of-the-art dereverberation models with supervision settings ranging from weakly-supervised to virtually unsupervised, relying solely on reverberant signals and an acoustic model for training. Most of the existing deep learning approaches typically require paired dry and reverberant data, which are difficult to obtain in practice. We develop instead a sequential learning strategy motivated by a maximum-likelihood formulation of the dereverberation problem, wherein acoustic parameters and dry signals are estimated from reverberant inputs using deep neural networks, guided by a reverberation matching loss. Our most data-efficient variant requires only 100 reverberation-parameter-labeled samples to outperform an unsupervised baseline, demonstrating the effectiveness and practicality of the proposed method in low-resource scenarios.

cs.SD↗
arXiv2026

S-PRESSO: Ultra Low Bitrate Sound Effect Compression With Diffusion Autoencoders And Offline Quantization

Neural audio compression models have recently achieved extreme compression rates, enabling efficient latent generative modeling. Conversely, latent generative models have been applied to compression, pushing the limits of continuous and discrete approaches. However, existing methods remain constrained to low-resolution audio and degrade substantially at very low bitrates, where audible artifacts are prominent. In this paper, we present S-PRESSO, a 48kHz sound effect compression model that produces both continuous and discrete embeddings at ultra-low bitrates, down to 0.096 kbps, via offline quantization. Our model relies on a pretrained latent diffusion model to decode compressed audio embeddings learned by a latent encoder. Leveraging the generative priors of the diffusion decoder, we achieve extremely low frame rates, down to 1Hz (750x compression rate), producing convincing and realistic reconstructions at the cost of exact fidelity. Despite operating at high compression rates, we demonstrate that S-PRESSO outperforms both continuous and discrete baselines in audio quality, acoustic similarity and reconstruction metrics.

cs.SD↗
arXiv2025

GLA-Grad++: An Improved Griffin-Lim Guided Diffusion Model for Speech Synthesis

Recent advances in diffusion models have positioned them as powerful generative frameworks for speech synthesis, demonstrating substantial improvements in audio quality and stability. Nevertheless, their effectiveness in vocoders conditioned on mel spectrograms remains constrained, particularly when the conditioning diverges from the training distribution. The recently proposed GLA-Grad model introduced a phase-aware extension to the WaveGrad vocoder that integrated the Griffin-Lim algorithm (GLA) into the reverse process to reduce inconsistencies between generated signals and conditioning mel spectrogram. In this paper, we further improve GLA-Grad through an innovative choice in how to apply the correction. Particularly, we compute the correction term only once, with a single application of GLA, to accelerate the generation process. Experimental results demonstrate that our method consistently outperforms the baseline models, particularly in out-of-domain scenarios.

cs.SD↗
arXiv2025

Harmonic-Percussive Disentangled Neural Audio Codec for Bandwidth Extension

Bandwidth extension, the task of reconstructing the high-frequency components of an audio signal from its low-pass counterpart, is a long-standing problem in audio processing. While traditional approaches have evolved alongside the broader trends in signal processing, recent advances in neural architectures have significantly improved performance across a wide range of audio tasks, In this work, we extend these advances by framing bandwidth extension as an audio token prediction problem. Specifically, we train a transformer-based language model on the discrete representations produced by a disentangled neural audio codec, where the disentanglement is guided by a Harmonic-Percussive decomposition of the input signals, highlighting spectral structures particularly relevant for bandwidth extension. Our approach introduces a novel codec design that explicitly accounts for the downstream token prediction task, enabling a more effective coupling between codec structure and transformer modeling. This joint design yields high-quality reconstructions of the original signal, as measured by both objective metrics and subjective evaluations. These results highlight the importance of aligning codec disentanglement and representation learning with the generative modeling stage, and demonstrate the potential of global, representation-aware design for advancing bandwidth extension.

cs.SD↗
arXiv2025

Is Phase Really Needed for Weakly-Supervised Dereverberation ?

In unsupervised or weakly-supervised approaches for speech dereverberation, the target clean (dry) signals are considered to be unknown during training. In that context, evaluating to what extent information can be retrieved from the sole knowledge of reverberant (wet) speech becomes critical. This work investigates the role of the reverberant (wet) phase in the time-frequency domain. Based on Statistical Wave Field Theory, we show that late reverberation perturbs phase components with white, uniformly distributed noise, except at low frequencies. Consequently, the wet phase carries limited useful information and is not essential for weakly supervised dereverberation. To validate this finding, we train dereverberation models under a recent weak supervision framework and demonstrate that performance can be significantly improved by excluding the reverberant phase from the loss function.

cs.SD↗
arXiv2025

PESTO: Real-Time Pitch Estimation with Self-supervised Transposition-equivariant Objective

In this paper, we introduce PESTO, a self-supervised learning approach for single-pitch estimation using a Siamese architecture. Our model processes individual frames of a Variable-$Q$ Transform (VQT) and predicts pitch distributions. The neural network is designed to be equivariant to translations, notably thanks to a Toeplitz fully-connected layer. In addition, we construct pitch-shifted pairs by translating and cropping the VQT frames and train our model with a novel class-based transposition-equivariant objective, eliminating the need for annotated data. Thanks to this architecture and training objective, our model achieves remarkable performances while being very lightweight ($130$k parameters). Evaluations on music and speech datasets (MIR-1K, MDB-stem-synth, and PTDB) demonstrate that PESTO not only outperforms self-supervised baselines but also competes with supervised methods, exhibiting superior cross-dataset generalization. Finally, we enhance PESTO's practical utility by developing a streamable VQT implementation using cached convolutions. Combined with our model's low latency (less than 10 ms) and minimal parameter count, this makes PESTO particularly suitable for real-time applications.

cs.SD↗
arXiv2025

Déréverbération non-supervisée de la parole par modèle hybride

This paper introduces a new training strategy to improve speech dereverberation systems in an unsupervised manner using only reverberant speech. Most existing algorithms rely on paired dry/reverberant data, which is difficult to obtain. Our approach uses limited acoustic information, like the reverberation time (RT60), to train a dereverberation system. Experimental results demonstrate that our method achieves more consistent performance across various objective metrics than the state-of-the-art.

cs.SD↗
arXiv2025

Soft Disentanglement in Frequency Bands for Neural Audio Codecs

In neural-based audio feature extraction, ensuring that representations capture disentangled information is crucial for model interpretability. However, existing disentanglement methods often rely on assumptions that are highly dependent on data characteristics or specific tasks. In this work, we introduce a generalizable approach for learning disentangled features within a neural architecture. Our method applies spectral decomposition to time-domain signals, followed by a multi-branch audio codec that operates on the decomposed components. Empirical evaluations demonstrate that our approach achieves better reconstruction and perceptual performance compared to a state-of-the-art baseline while also offering potential advantages for inpainting tasks.

cs.SD↗
arXiv2025

Désentrelacement Fréquentiel Doux pour les Codecs Audio Neuronaux

While neural-based models have led to significant advancements in audio feature extraction, the interpretability of the learned representations remains a critical challenge. To address this, disentanglement techniques have been integrated into discrete neural audio codecs to impose structure on the extracted tokens. However, these approaches often exhibit strong dependencies on specific datasets or task formulations. In this work, we propose a disentangled neural audio codec that leverages spectral decomposition of time-domain signals to enhance representation interpretability. Experimental evaluations demonstrate that our method surpasses a state-of-the-art baseline in both reconstruction fidelity and perceptual quality.

cs.SD↗
arXiv2025

Melody-Lyrics Matching with Contrastive Alignment Loss

The connection between music and lyrics is far beyond semantic bonds. Conceptual pairs in the two modalities such as rhythm and rhyme, note duration and syllabic stress, and structure correspondence, raise a compelling yet seldom-explored direction in the field of music information retrieval. In this paper, we present melody-lyrics matching (MLM), a new task which retrieves potential lyrics for a given symbolic melody from text sources. Rather than generating lyrics from scratch, MLM essentially exploits the relationships between melody and lyrics. We propose a self-supervised representation learning framework with contrastive alignment loss for melody and lyrics. This has the potential to leverage the abundance of existing songs with paired melody and lyrics. No alignment annotations are required. Additionally, we introduce sylphone, a novel representation for lyrics at syllable-level activated by phoneme identity and vowel stress. We demonstrate that our method can match melody with coherent and singable lyrics with empirical results and intuitive examples. We open source code and provide matching examples on the companion webpage: https://github.com/changhongw/mlm.

eess.AS↗
arXiv2025

Episode-specific Fine-tuning for Metric-based Few-shot Learners with Optimization-based Training

In few-shot classification tasks (so-called episodes), a small set of labeled support samples is provided during inference to aid the classification of unlabeled query samples. Metric-based models typically operate by computing similarities between query and support embeddings within a learned metric space, followed by nearest-neighbor classification. However, these labeled support samples are often underutilized--they are only used for similarity comparison, despite their potential to fine-tune and adapt the metric space itself to the classes in the current episode. To address this, we propose a series of simple yet effective episode-specific, during-inference fine-tuning methods for metric-based models, including Rotational Division Fine-Tuning (RDFT) and its two variants, Iterative Division Fine-Tuning (IDFT) and Augmented Division Fine-Tuning (ADFT). These methods construct pseudo support-query pairs from the given support set to enable fine-tuning even for non-parametric models. Nevertheless, the severely limited amount of data in each task poses a substantial risk of overfitting when applying such fine-tuning strategies. To mitigate this, we further propose to train the metric-based model within an optimization-based meta-learning framework. With the combined efforts of episode-specific fine-tuning and optimization-based meta-training, metric-based models are equipped with the ability to rapidly adapt to the limited support samples during inference while avoiding overfitting. We validate our approach on three audio datasets from diverse domains, namely ESC-50 (environmental sounds), Speech Commands V2 (spoken keywords), and Medley-solos-DB (musical instrument). Experimental results demonstrate that our approach consistently improves performance for all evaluated metric-based models (especially for attention-based models) and generalizes well across different audio domains.

cs.LG↗
arXiv2025

Diff-TONE: Timestep Optimization for iNstrument Editing in Text-to-Music Diffusion Models

Breakthroughs in text-to-music generation models are transforming the creative landscape, equipping musicians with innovative tools for composition and experimentation like never before. However, controlling the generation process to achieve a specific desired outcome remains a significant challenge. Even a minor change in the text prompt, combined with the same random seed, can drastically alter the generated piece. In this paper, we explore the application of existing text-to-music diffusion models for instrument editing. Specifically, for an existing audio track, we aim to leverage a pretrained text-to-music diffusion model to edit the instrument while preserving the underlying content. Based on the insight that the model first focuses on the overall structure or content of the audio, then adds instrument information, and finally refines the quality, we show that selecting a well-chosen intermediate timestep, identified through an instrument classifier, yields a balance between preserving the original piece's content and achieving the desired timbre. Our method does not require additional training of the text-to-music diffusion model, nor does it compromise the generation process's speed.

cs.SD↗